working paper
department
of economics
ECONOMIC EFFECT OF
TELEVISION-NETWORK PROGRAM "OWNERSHIP"
Robert W. Crandall
Number 50
<hk~
C^"C^\
February 1970
massachusetts
institute of
technology
50 memorial drive
Cambridge, mass. 02139
:.T£CH.
FEB
23
1970
DEWEY LIBRARY
ECONOMIC EFFECT OF
TELEVISION-NETWORK PROGRAM "OWNERSHIP"
by
Robert W. Crandall
Number 50
'hk-
C^°VA
A
February 1970
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium Member Libraries
http://www.archive.org/details/economiceffectofOOcran
Economic Effects of Network "Ownership" of Television Programs
Robert W. Crandall*
It does not require great perception to detect widespread dissatisfac-
tion with the quality of network television programming in the United States.
Criticism of program quality and diversity seems to increase with time, but the
programs change very little.
This is most discouraging in view of the fact
that television is subject to social control; but, given the diversity of opinions and political pressures which bear upon the Federal Communications Com-
mission, it is perhaps not surprising that the Commission is unable to pursue a
clear, consistent policy which might improve television-industry performance.
Recent changes in policy and personnel at the F.C.C. give rise to some
optimism that significant improvements in programming may yet be forthcoming.
These changes may be delayed if the Commission is to continue to pursue a regulatory proposal advanced by its Office of Network Study four years ago.
In 1965, the F.C.C. published a Proposed Rule which would limit net-
work prerogatives in licensing their prime-time entertainment programs.
This
Rule lay dormant for three years, but 1968 found the Commission interested once
more in considering its adoption.
Unfortunately, the Rule has been virtually
* Assistant Professor of Economics, Massachusetts Institute of Technology.
My initial interest in this subject developed while I was a consultant for the
Antitrust Division of the United States Department of Justice. Data utilized in
this study were supplied by the networks, and I am grateful to these companies for
their cooperation. Of course, the opinions contained herein are solely my own and
do not reflect the views or opinions of the network companies or the Department of
Justice. Statistical computations were performed by Susan Shea.
1. U.S. Federal Communications Commission, Notice of Proposed Rule-Making ,
In the Matter of Amendment of Part 73 of the Commission's Rules and Regulations
with Respect to Competition and Responsibility in Network Television Broadcasting,
Docket 12782, 1965.
2. See "Playback on 50-50 Proposal," Variety ,
September 25, 1968.
ignored by students of public regulation, and its attributes have been debated
only by those parties directly affected by it.
The Commission has drafted the Proposed Rule in order to improve pro-
gram quality, but careful analysis of industry structure and performance leads
to the conclusion that the Rule will have little effect upon quality or divers-
ity of programming.
Indeed, the Rule will have little effect upon any important
parameter other than network profits, but continued study and consideration of
this proposal is a serious drain upon the resources and energies of the Commis-
sion which might be devoted to more substantial issues in the regulation of
television.
I.
The Source of the Problem:
Network functions
local stations' time.
.
Network Activities
Basically, a network is no more than a broker of
The network company secures contracts with a number of
geographically-dispersed stations for access to their frequencies, and in turn
the network sells to advertisers time in which to convey programs and attached
commercial messages over the entire 'network' of stations.
provide cable or microwave interconnection, nor
need not own stations,
produce programs.
The network company
It merely must maintain a brokerage office to bring national
advertisers, the buyers, into contact with dispersed licensed television stations, the sellers, for the purpose of utilizing the stations' broadcasting time.
Of course, the three major networks today do not resemble the simple
model of a broker.
Networks own television stations, produce programs, and buy
other programs for resale to advertisers, but interconnection services are still
purchased from A.T.ST.
I
shall not be concerned with the reasons for network
3. Even this function may be integrated into network activities if network
managers are unable to persuade the F.C.C. or A.T.§T. to lower the fees for microwave circuits leased by the networks. As this paper is being written, the network
companies are suggesting that they be given permission to establish their own
domestic satellite system for interconnection of their affiliated stations.
ownership of stations, but it is probable that the sole limit to network inte-
gration of this sort is to be found in F.C.C. regulations, which limit ownership of stations by a single organization to five in the VHF band.
The bargain-
ing pressures of bilateral oligopoly would undoubtedly press the networks to
expand their ownership of local stations in the absence of F.C.C. regulation.
My primary concern is with the networks
program production and program brokerage.
1
backward integration into
These functions are now under F.C.C.
attack for a variety of reasons.
As licensees of five stations each, the three networks are subject to
the requirement (contained in the Federal Communications Act) that they broad-
cast in the public interest.
as national news,
4
This requires the transmission of programs such
local news, documentaries, local features, and other generally
nonremunerative programs.
As a result, each network has found it necessary to
develop its own news and public service programming facilities and staff.
Pre-
sumably, these functions could be contracted out, but the networks undoubtedly
feel that control is more easily maintained through their own organization.
In
addition, maintenance of their own facilities probably allows greater flexibility
in adapting to unforeseen news events and in assuring fairness in covering vari-
ous editorial points of view.
While it is not surprising to find networks producing their own news
and public-service programming, similar integration into entertainment program-
ming is more difficult to explain.
There would seem to be little reason for
networks to produce their own westerns or situation comedies for continuing
series or to produce motion pictures.
Show business would seem to be somewhat
4. Networks, as such, do not have statutory responsibility for broadcast
matter.
See Television Network Program Procurement , Second Interim Report by
the Office of Network Study (U.S. Federal Communications Commission, 1965), p.
200.
(Hereafter this volume will be referred to as the Second Interim Report .)
4
alien to a broker of station time, but a glance at the history of the industry
will show that networks have always produced some of their own entertainment
programs although the past ten years have witnessed a striking decline in
network-produced shows.
As a substitute for their own productions
,
networks
have turned to outside producers, but in so doing the networks have undertaken
a new function -- the brokerage of programs.
Program sources
programming are
(3)
(1)
.
The three basic sources of network entertainment
advertisers, (2) independent producers or "packagers," and
the networks' own production facilities.
1.
Advertisers.
has been the advertiser.
Historically, an important source of prime-time series
An advertiser may arrange to procure his own program
series from an independent production company and bring this series to a net-
work for exhibition.
In such "advertiser-supplied" series, the advertiser
merely pays the network a time rate for access to its interconnected stations,
and the network does not intercede in program production details.
In practice,
an advertiser will usually seek prior network approval before investing heavily
in a new series since network exhibition is by no means assured for every series
which an advertiser may wish to purchase.
An important aspect of this source of programming is the absence of
network financial participation in the initial development of the series or in
subsequent exploitation of episodes which are shown on the network.
The adver-
tiser and the producing company share in the risks and the financial rewards
from such series, and the network's only compensation is from sale of time for
the exhibition of the series during its first run.
5. Usually, the advertiser and network enter into a one-year contract
This "first -run" agreement does not constifor the exhibition of such a series.
tute "network ownership" in the Commission's view, but apparently a five-year
contract between supplying advertiser and the network would constitute network
"ownership" of first-run rights.
2.
Independent producers.
Instead of relying upon advertisers to ar-
range for their own entertainment programming, networks have begun to contract
directly with independent producers or "packagers."
The networks then sell
commercial minutes in these "packager-licensed" programs, varying the charge
with the expected audience for the programs.
Advertisers may purchase one or
more commercial minutes in each of a large number of such packager-licensed
shows as an alternative to supplying their own programs to the network.
7
The network's role in packager-licensed programs is significantly different from that in advertiser-supplied shows.
Usually, the network supplies at
least part of the risk capital in developing the program format, producing the
pilot, and producing the first few episodes.
In addition, the network usually
obtains the right to exhibit the program series over its facilities for at least
five years while the usual agreement between advertiser and network for adver-
tiser-supplied programs is only one year.
Finally, the network and packager
negotiate for shares in all ancillary revenues which may emerge from a successful
series.
Thus, a network exposes itself to considerable risk in procuring its
programs directly from independent packagers, but in return it may reap very
large returns from network exhibition and ancillary rights of the successful series,
3.
Network production.
Each network is capable of producing entertain-
ment programming of its own although such activity has been diminishing in recent
years.
Obviously, in network-produced programs, the network underwrites the
entire risk and reaps all of the consequent gains (or suffers all of the losses)
6. "Packagers" are program-supplying companies.
"producers" are creative agents employed by packagers.
In the industry parlance,
7. The number of commercial minutes per hour of prime-time programming
is not controlled by the F.C.C., but an industry tradition limits the number to
six for all prime-time programs except motion pictures, in which there are seven.
Advertisers may purchase all of the commercial minutes in any series if they
choose, but this form of complete sponsorship of an individual series is declining as advertisers spread their messages among a larger number of series.
.
6
from the successful (or unsuccessful) shows.
As in packager-licensed pro-
grams, time is sold to advertisers in these shows by commercial minute, and
the price is a function of the expected number of viewers.
Since 1957, the advertiser-supplied program has been declining in
relative importance and has almost disappeared as networks have turned to
licensing their shows directly from independent packagers.
In 1957, advertisers
supplied 36.1 per cent of all prime-time hours of network entertainment programming.
Networks produced 23.9 per cent of their own entertainment hours,
licensed 39.3 per cent directly from packagers, and secured another 0.7 per cent
in a combination of these ways.
By 1968, the advertiser-supplied show had de-
clined to only 3.4 per cent of prime -time entertainment hours while packagerlicensed programs had risen to a 91.2 per cent share.
Networks' own production
of these programs declined to only 4.8 per cent of prime-time hours; the remaining 0.6 per cent are obtained by combination of these means.
Accompanying this trend toward direct licensing by the networks has been
an expansion in their role in the financing, development, brokerage, and subse-
quent exploitation of programs in all markets.
It is this expanded role which
alarms the F.C.C. and which requires further analysis.
Program brokerage .
Central to the issue posed by the Federal Communica-
tions Commission is the networks' program-brokerage function.
It is this act
of intermediation which the Commission sees as severely constraining program
packagers and reducing program quality.
Actually, the network is both a broker and dealer of entertainment programs.
It
contracts with packagers for nearly all of its prime-time entertain-
ment series and resells time in these programs to advertisers.
While it need
8. These data may be found in Television Program Production, Procurement ,
(Hereafter, referred to as the
Distribution, and Scheduling A. D. Little, 1969.
Little Report [1969]), p. 5. Prime time is defined as the hours between 6:00 p.m.
and 11:00 p.m. , Eastern Time
,
not make a financial commitment in programs before advertisers indicate their
preferences, the network finds it convenient to do so in order to assemble its
season's programs far in advance of their first appearance on the air.
There-
fore, the network is more than a mere broker of programs for it provides invest-
ment capital for program development and commits itself to procurement of programs in advance of sales to advertisers.
The brokerage function of a network has developed slowly over the past
decade.
In previous years, networks sold time in programs to advertisers on a
weekly or alternate-week basis for those programs which it produced or procured
from outside suppliers.
Advertisers might also bring programs to the networks
through their advertising agencies, but networks generally dissuaded advertisers
from sharing advertising time with more than one other advertiser in any series.
In 1957, only 14 per cent of all prime-time entertainment series were sponsored
by more than two advertisers.
9
Beginning in the early 1960s, networks began to sell advertising time
in series on a "participating" basis, allowing an advertiser to purchase a few
one-minute commercial announcements in programs rather than a full season's
commitment to a given program.
This development proceeded hand in hand with
the continuing trend toward packager-licensed programs, for networks could more
easily develop such flexibility in sale to sponsors by first obtaining their
programming from suppliers.
By 1967, networks were selling commercial minutes
to three or more sponsors in series accounting for over 90 per cent of their
In fact, series sponsored by twenty-five or more separate
prime-time hours.
advertisers occupied more than half of the prime-time entertainment hours on
the networks in 1967.
9.
10.
Little Report (1969), p. 21.
Ibid.
,
p.
17.
Whether this change derived from advertisers' wishes or was instigated unilaterally by the networks is difficult to determine and irrelevant
to tiie analysis which follows.
11.
Ibid.
8
Program syndication
.
All network programs which are recorded on film
or videotape have extra-network revenue potential from resales to independent
domestic television stations and foreign television networks or stations.
Sales
of these "reruns" of network programs are referred to as "syndication" sales.
In addition,
the successful program may be exploited by attaching some of its
copyrighted features to merchandise.
Licensing of copyrighted features of these
programs in such a fashion gives rise to "merchandising" revenues in the parlance
of the industry.
Since networks now obtain nearly all of their entertainment programs
directly from independent packagers, they may negotiate ex ante for a share of
the revenues from any of the above sources.
In practice, merchandising revenues
are not very important, amounting to less than 5 per cent of the net revenues
from syndication to the networks in 1967.
Therefore,
I
shall ignore this rather
heterogeneous merchandising market in the discussion which follows.
Networks and packagers may obtain income from syndicating network programs in two ways.
First, either the networks or the producers may perform the
actual distribution and promotion of the syndicated series for a percentage of
the gross revenues.
These syndication fees vary with the size of the transaction
involved and are often different for domestic and for foreign sales; but, as with
most other parameters involved, the fees are agreed upon by packager and network
before the program is ever shown on the network.
As the networks have obtained less of their programming from advertisers,
they have had greater opportunity to bargain for the rights to engage in the
syndication of programs.
But the percentage of total network entertainment hours
in which networks have the rights to future syndication has not risen.
In 1957,
networks had these rights in 30.2 per cent of the entertainment hours currently
in network exhibition.
This share rose to a high of 35.0 per cent in 1964, but
it subsequently declined to 25.9 per cent in 1968.
12
In practice,
the networks
claim that they usually forego these rights if the packager has his own syndication organization although some exceptions to this practice do exist.
A second source of revenue from syndication derives from "profit shares"
These profit shares are obtained from the net revenues from
in the programs.
syndication which are merely gross revenues less syndication fees, production
"overages," and other costs.
Overages are reimbursements to packagers for (un-
expected) expenses incurred in production of the original episodes.
Besides
the networks and the packagers, participants in profits from syndication may
include actors, directors, producers, or other creative personnel.
In addition,
fees for members of theatrical guilds must be deducted before the network and
packager are able to claim their shares.
The percentage of all entertainment series currently being exhibited on
network television in which networks have a share of syndication profits has
risen substantially since 1957.
In that year, networks had a profit share from
domestic syndication in 36.5 per cent of prime-time hours and from foreign syndication in 37.2 per cent of prime-time entertainment hours.
By 1966, these shares
had risen to 67.8 for both markets, and they stood at 58.1 and 59.5, respectively,
in 1968.
Of those shows licensed directly to the networks by packagers, the
networks obtained profit shares in nearly 60 per cent in 1968.
13
According to network data reproduced in the Little Report , the three
networks' net gain from syndication rights rose from $893,000 in 1960 to
$3,509,000 in 1967 while net gain from syndication profit shares rose from
$1,054,000 to $4,282,000.
These "net gains" reflect gross revenues less all
costs incurred, including an unspecified allocation of joint administrative costs,
Little Report (1969), p. 46.
F.C.C.'s Office of Network Study.
12.
These data may also be obtained from the
13.
Little Report (1969), pp. 48 and 54.
14.
Ibid.
,
pp. 64-65.
.
10
III.
The F.C.C.'s Complaint
In its Proposed Rule and the background studies which preceded it, the
F.C.C. has taken a dim view of the above developments.
It sees them as inimical
to competition in program supply, syndication, and network broadcasting.
Specifi-
cally, the Commission argues that:
a) The
vitality of competition in program supply has been reduced by the
concentration of buyers of programming occasioned by the networks' shift to direct
licensing of their prime-time series.
b)
Competition has also been reduced in syndication markets as the net-
works now have "ownership" interests in all but four per cent of network entertainment program hours.
c) The
movement to packager-licensed programming has increased network
control over programs
d) Networks have used their monopsony power in licensing programs to
obtain a financial interest in these programs, refusing to grant network exhibition for those programs in which the packager insists upon all syndication and
merchandising rights and profit shares.
The Commission states in the concluding section of the Notice of Proposed
Rule-Making that:
"We proposed to encourage and increase competitive forces-- both
creative -and economic -- in television production and procurement
through limitations on the capacity of network corporations to
confine network schedules to programs in which they have financial
and proprietary interests and through divorcement of networks from^
domestic syndication and, to some extent, foreign distribution."
15.
U.S. Federal Communications Commission, op. cit.
,
p.
18,
11
The Commission's Proposed Rule stipulates that:
16
i.
Networks be forbidden to obtain syndication rights or profit shares
in any program not produced by them.
Networks be forbidden from distributing programs which they produce
ii.
to domestic syndication markets or to retain profit shares from domestic syndicaThey would be forced to sell their rights to domestic syndication to antion.
other entity, but they would be permitted to distribute their own programs in
foreign markets.
During prime time, networks be required to obtain at least 50 per
iii.
cent of their entertainment program hours from sources other than their own production facilities or direct licensing from packagers.
(In terms of the discussion in Section I, above, this would require that at least 50 per cent of
programs be advertiser-supplied or be placed through some other broker.)
In short, the network's program brokerage function is to be contracted
by approximately fifty per cent for entertainment programs, and it is to be
forbidden from all participation in domestic syndication revenues and nearly
In this way, program diversity is
all access to foreign syndication markets.
to be increased, the supply of programs in syndication is to be expanded, and
the number of U.1I.F. stations will increase as the amount of programming avail-
able to them in syndication markets expands.
As a result, the likelihood of a
fourth network is enhanced because of the increase in U.H.F. outlets in markets
now served by three or less stations.
The remaining sections of this paper will be devoted to a thorough exam-
ination of the F.C.C.'s charge and the resulting provisions in the Proposed Rule.
Before proceeding, however, a brief digression on the source of network power is
required.
III.
Concentration in Network Television
-
The Outcome of F.C.C. Regulation
One of the widely accepted notions in economics is that markets in which
there exists "natural monopoly" require regulation by an outside authority in
16. The text of the Proposed Rule may be found in the Notice Appendix A,
or in A. D. Little, Television Program Production, Procurement, and Syndication ,
1966, Vol. I, Appendix A.
,
12
order to assure that prices are reasonably close to (socially-necessary) incremental costs.
Communications markets have come to be regulated for quite another
reason, however: the absence of property rights in the electromagnetic spectrum.
In fact,
it is regulation which has created monopoly power in one of the most
important communications markets -- that of television broadcasting.
Were it
not for F.C.C. policy, there might be a larger number of competing national broad-
casting companies.
It is the F.C.C. which allocates the electromagnetic spectrum
variety of uses, including television.
among a
In its allocations plan which was adopted
in 1952, the Commission sought to achieve maximum geographic dispersion of tele-
vision stations despite complaints by the smallest of the four existing networks,
Du Mont, that such a plan would force it to exit from network broadcasting.
17
The allocation of frequencies by the F.C.C. was accomplished without an
analysis of the current or prospective opportunity costs of spectrum space.
Television has been granted most of the space between 54 megacycles and 210
megacycles in the V.H.F. band, and all of the space between 470 megacycles and
890 megacycles in the U.H.F. band.
The Commission's delay in allocating the
U.H.F. band has caused television stations operating at these frequencies to
suffer a competitive disadvantage since millions of receivers were constructed
without U.H.F. capability.
Whether the Commission has erred in granting too little or too much
spectrum space to television transmission cannot be answered in this paper since
neither the Commission nor other students of communications have been able to
estimate the incremental opportunity costs of the spectrum or of satisfactory
substitutes for the spectrum.
But the Commission has overtly limited the use
of substitutes in television broadcasting by impeding the development of pay
17. Complete details of the allocation plan may be found in F.C.C,
"Sixth Report on Television Frequency Allocations," 1952.
13
television and community antenna systems (CATV)
18
'
.
In short, the F.C.C. has
until this time limited competition in television broadcasting by establishing a limit on the number of conduits through which television signals may be
distributed.
It is the geographical distribution of television stations in the V.H.F.
band which is most important in explaining current market structure in network
television.
Due to problems of adjacent channel interference, the maximum
feasible number of V.H.F. stations in any single location is seven (Channels 2,
4,
5,
7,
9,
11, and 13), but every location cannot have seven stations due to
cochannel interference.
Given the distribution of population in the United
States, a maximum allocation of, say, seven stations to New York and seven to
Boston would make it difficult to provide many V.H.F. stations to Albany, Philadelphia, Hartford, Providence, Springfield (Mass.), and Manchester (N.H.).
Re-
duction in the power allowed to the New York and Boston stations might alleviate
this problem, but then there would be large gaps in reception between the cities,
or much higher antenna costs imposed upon viewers in these areas.
In basing its allocation plan upon the seemingly equitable idea that
each community should have a station wherever possible, the Commission granted
very large cities between three and seven outlets, but where conflicts existed,
it chose to allow smaller communities at least one station and to reduce the
number of outlets in the large cities.
Thus, Toledo (Ohio), Hartford-New Haven,
Dayton (Ohio), Lansing (Michigan), and Charlotte (N.C.), for instance, each have
but two V.H.F. outlets although each market is among the largest thirty in the
nation.
In only fourteen markets, comprising less than 40 per cent of the
nation's television homes, are there four or more commercial V.H.F. stations.
19
18. The F.C.C. has just announced a change in its CATV policy, permitting
and even requiring the origination of programming by Community Antenna systems.
This is a welcome change which may lead to greater competition and diversity in
television broadcasting.
19. Television Digest,
inc., Television Factbook,
1963.
14
In the largest 50 television markets, 76 per cent of television homes are able
to receive a fourth signal from either U.H.F. or V.H.F. band, but in smaller
markets, comprising nearly half of the nation's television homes, this percentage is much smaller.
20
In only twenty-two of the markets ranked from 50 through
230 on the basis of weekly circulation, is there a fourth television station in
either U.H.F. or V.H.F.
stations.
21
In many of these markets, there are less than three
While some homes in these smaller markets may be able to receive an
independent fourth station from nearby large television markets and others may
do so through CATV systems, it would be surprising if more than 60 per cent of
the nation's television homes were within reach of a fourth commercial station
which is not a primary affiliate of one of three existing networks.
22
It is this allocations policy which has limited the number of national
networks to three, resulting in limited program diversity and quality.
23
'
That a
fourth network cannot hope to compete with the three major firms currently in
existence is quite clear when one examines the nature of competition in network
broadcasting.
20.
Little Report (1969), p. 151.
21. Television Digest, Television Factbook ,
1968.
22. Overlapping markets will often find some homes capable of receiving
more than three commercial stations, but most often these stations are not a source
of variety in programming since they are affiliated with the same three networks.
Given that most homes within the range of a given station in these markets cannot
receive more than three signals, a fourth network will be unable to obtain outlets
there.
23. The conclusion that fewness of broadcast services leads to lack of
diversity in programming is quite generally accepted by students of the industry.
See Peter Steiner, "Program Patterns and Preferences and the Workability of Competition in Radio Broadcasting," Quarterly Journal of Economics , May 1952, pp. 194223; Jerome Rothenberg, "Consumer Sovereignty and the Economics of Television Programming," Studies in Public Communication , Number 4, Autumn 1962, pp. 45-54; John
J. McGowan, "Competition, Regulation, and Performance in Television Broadcasting,"
Washington University Law Quarter ly, 1967, No. 4, pp. 499-519.
24. A more rigorous examination of the nature of network competition and
program diversity requires the use of mathematical exposition which is too intricate for the purposes of this paper. See Robert W. Crandall, "Television Network
Competition and Program Diversity," unpublished manuscript, 1969.
15
.
Network managers cannot increase the number of prime-time hours of
programming which they send to their interconnected affiliates once they have
encountered the constraint of 28 hours per week.
Competition can then only
take the form of quality variations in programs.
Specifically, a model of the
profit-maximizing network tells us that each network will choose its programs
for each hour of prime time in such a manner as to maximize the difference
between the additional value of the program, measured in additional advertising
revenues, and the cost of the program.
25
"
This incremental profit may be repre-
sented quite simply as:
MP = MR
where MP
MA
=
•
MA
-
p
incremental profit from a given hour of network broadcasting.
addition to the network's viewer hours for the programming
period (in our case, a single week).
MR = incremental revenue per additional yiewer hour.
p = the price of the program episode paid by the network to the packager.
= the
But MA is itself a function of p
-- the price of the program -- since a network
may entice additional viewers to its program during a given hour by increasing
its program expenditures.
This may take the form of securing high-rent perform-
ers, using special video effects, procuring scripts from well-known writers, or
utilizing other costly production techniques.
For a given type of programming,
a network is less likely to expend large sums on programs if it has a complete
monopoly in television broadcasting since it can only increase its audience
through attracting viewers who otherwise would not be watching television at all,
If there are rival oligopolists, however, and if the incremental profitability
of a given type of programming is high for a specific price of programming, it
is likely that a network will seek to attract viewers from the others' programs
25. I ignore the costs of interconnection and affiliate compensation as
well as other overhead costs in this discussion because they do not vary among
different types of programs broadcast by the network. Unless the marginal
profitability of the programs broadcast by a network are sufficient to cover these
fixed costs, the network will not broadcast a full 28 hours per week. At present,
this possibility looms only for A. B.C., the least profitable of the three networks,
.
16
by adding to its program expense.
This procedure will continue until each
network feels that adding to the cost of the given program will not result in
an equivalent increase in its marginal value product (MR*MA)
The disadvantage faced by an entrant which cannot attract a roster of
affiliates capable of reaching a very large percentage of the nation's television homes may be depicted in Figure
The smaller the entrant's number of
1.
affiliates, the smaller will be his potential audience for any given hour of
programming (MA)
.
Given an array of possible program expenditures upon a parFigure
1
Profitability of One Hour of a Given Type of Network Programming
MA
•
(MA«
MR
existing firms
MR
P«
potential entrant
ticular type of programming,
the entrant's MR»MA curve will be correspondingly
lower than that of his existing rivals.
programs of price p*
,
If these established networks choose
they will receive a gross margin of (MR«MA),
-
p* from which
to defray their costs of interconnection, affiliate compensation, and overhead
expense.
The entrant may not be able to cover his program costs, much less other
fixed costs, from this type of programming if he can reach but 60 or 70 per cent
26. The final equilibrium depends upon the behavioral assumptions underlying the oligopoly model.
It is conceivable that networks might arrive at the
joint-profit maximizing program structure subject to the constraint of no side payments, but this outcome is extremely unlikely due to the difficulty of policing
program decisions and responding to rivals' successful decisions rapidly.
27. At any time a network may choose from a number of different program
"types," but whatever the entrant's choice, its incremental revenue will be less
than that of its rivals from similar programs. The discussion above is limited to
a single type of program as a matter of graphical and expositional convenience.
17
of the nation's television homes.
Unless rivalry among networks is so limited
as to result in very large gross profit margins, it is unlikely that a fourth
network could survive.
While data on prices and costs of entertainment series
are scarce, it is doubtful that the outcome of current network rivalry allows
profit margins on prime-time series of more than 10 to 15 per cent after the
deduction of all fixed costs.
28
'
A fourtli network would probably reduce this
margin for existing networks, and it would hardly be able to compete with these
networks without a complement of affiliates capable of reaching, say, 85 to 90
per cent of the nation's television homes.
Since such circulation is impossible
given the F.C.C. allocations plan, a fourth network cannot hope to compete with
the three major firms by offering programs of similar type.
Might there be programs which a small network could exhibit at a profit
-- programs which existing
networks shun simply because more expensive, mass-
appeal programming leads to larger net profits?
This seems exceedingly doubtful
for a number of reasons.
28. F.C.C. Broadcast Financial Data show a rather constant 8.5 per cent
net profit margin before taxes for the networks in 1964-66, but there is no way
of assessing the validity of such data given the necessity of intra-firm divisions
It is undoubtof costs and revenues among owned stations and network operations.
edly true that the net profit margins on entertainment programming are much higher
since networks presumably obtain very low (perhaps even negative) profit margins
from news and public-service broadcasting. Given new program costs of $160,000
per hour in 1967 and median revenues of $210,000 per prime time entertainment hour
in the fall of 1967, it is possible that network profit margins were no more than
10 per cent in even the most profitable of hours.
Since approximately two-thirds
of all prime-time programs are held over from the previous seasons, the average
program costs to networks may be estimated at $140,000 per hour in 1967 in the absence of escalation clauses. A maximum estimate of gross margins for fall programs
would therefore be $210,000 minus $140,000, or $70,000, from which affiliation payments and interconnection costs must be paid. These costs may be estimated at
$45,000 - $50,000 per prime-time hour for 1967, leaving $20,000 - $25,000 in residual "profits" from which administrative costs must be deducted. The resulting
net profit margin is clearly less than 15 per cent and may even be less than 10
For instance, there are no
per cent, but the calculations are only approximate.
data on the price of repeat episodes for these programs which are shown in spring
and summer.
These repeat episodes bring smaller audiences and therefore lower
gross revenues -- a median value of $190,000 per hour in April, 1968, and $130,000
per hour in July, 1968 ( Little Report [1969] , p. 45) -- but repeat episodes may be
much less costly to networks than the first run of these episodes. Therefore, it
.
18
First, it is difficult to imagine a program which would allow a
fourth network to cover its fixed costs of interconnection, which may be as
high as $100,000 per week, and its affiliation costs, but which would not
be enormously profitable to a network which has the capacity to reach a much
larger home audience.
Secondly, the types of programs which might fit into
such a category are likely to be "talk" shows or films of old sports events,
documentary films, or the like which are available now in syndication markets.
The formation of a network to simulcast such fare seems most unlikely.
29
Finally,
recent attempts to produce inexpensive dramas of the "soap opera" variety for
original use on independent stations throughout the country have not met with
great success despite a market which potentially encompasses all of the stations which a fourth network could attract.
30
The notion that the paucity of affiliates is the limiting factor in
network broadcasting is amply borne out by a number of independent observations.
From 1948 through 1955, the DuMont network competed with its larger rivals, but
it was forced to exit when the F.C.C. allocations plan made it impossible for
DuMont to gain access to sufficient affiliates.
31
At the present time, it is widely acknowledged that A. B.C. is the
least profitable of the three major networks, and A. B.C. has even claimed that
would be difficult to measure precisely the magnitude of net profit margins in
prime-time entertainment programming, the reader should note that 10-15 per cent
net margins may reflect extremely high profit rates (as a percentage of capital
invested) in this industry given the relatively high sales/capital ratio.
29. The Little Report (1969), p. 79, estimates that "talk" and "talkvariety" programs account for over 50 per cent of all program hours produced
specifically for nonnetwork exhibition (the"first-run syndication" market)
Adventure, mystery, and comedy shows account for only 10 per cent of such
programming, reflecting the difficulty of filming these programs for only the
nonnetwork market.
30. See Variety , September 24, 1969, p. 39, for a description of the difficulties encountered by producers of this type of programming.
31. Se the testimony of A.B. DuMont in Hearings , Subcommittee on Communications, Senate Interstate and Foreign Commerce Committee, 83rd Congress,
2nd Session, 1954, for a discussion of the problems which faced the DuMont net-
work.
19
its network television activities are undertaken at a loss.
32
A large part
of this difference in performance between A. B.C. and its rivals can be explained
by its roster of affiliated stations.
In those markets in which there are three
or more affiliates, A. B.C. typically has the least attractive outlet.
The gross
network time rate charged by a network for each station for advertiser-supplied
programs and from which the network calculates its affiliate compensation may be
taken as a rough measure of the potential audience capability of the station.
In the three-or-more station markets, A.B.C.'s affiliates' gross network time
rates are nearly 12 per cent below those of N.B.C. affiliates and 9 per cent
below those of C.B.S. affiliates.
33
A similar disadvantage undoubtedly carries
over into one- and two-station markets in which A. B.C. is less likely to obtain
clearance for its programs than are C.B.S. and N.B.C.
In its decision on the proposed merger between A. B.C. and I.T.5T. in
32.
1967, the Federal Communications Commission accepted the argument that A.B.C.'s
network television operations were operating at a loss during the 1963-1966
period. This loss increased from $4.6 million in 1963 to $9.0 million in 1966,
while the combined before-tax profits of C.B.S. and N.B.C. network activities
were rising from $61.0 million to $87.7 million. Of course, these data reflect
cost and revenue allocations between network activities and owned-and-operated
station operations. A. B.C. profits from their owned-and-operated television
stations were more than sufficient to offset A. B.C. network losses in each year.
(See P&F Radio Regulation Reports, 10 RR 2d 289 ff , June 22, 1967.)
.
33. The sum of the gross hourly rates for 101 three-station markets,
based upon the latest rate cards available for publication in the 1968 Television
Factbook , is $108,425 for A. B.C., $119,000 for C.B.S., and $123,390 for N.B.C.
It is reasonable to assume that differences in rates among stations in the same
market reflect differences in audience potential due to assigned frequency differences, differences in maximum broadcast wattage allowed by the F.C.C., or
variance in station management.
It is interesting to note that A.B.C.'s gross revenues from its television
network were $161 million in 1961 while C.B.S. and N.B.C. realized an average of
$182.8 million in the same year. Thus, A. B.C. obtained nearly 12 per cent less
gross revenue from networking than its larger rivals -- a magnitude almost identiThe F.C.C. is
cal to its affiliation disadvantage in the three-station markets.
not unaware of this problem, and it once considered increasing the number of
V.H.F. allocations in some cities to alleviate it.
But after considerable delibera(See F§P Radio Regulation Reports,
tion, this idea was dropped in January 1960.
loc. cit.)
20
A recent entry attempt by the United Television Network is further evi-
dence of the importance of outlets which are capable of reaching all television
homes.
In May,
1967, United began broadcasting nationally by exhibiting one
program per day at 11:30 P.M.
-
1:30 A.M.
34
This decision was made because at
that time period C.B.S. had no network service and C.B.S. affiliates could be
attracted by United.
Despite a roster of 106 affiliates, United suspended
operations after only one month of broadcasting.
fixed costs
It simply could not cover
of network operation from the revenues from only 10 hours of
programming per week during a period in which total audiences are small.
did it not expand its program schedule to prime-time hours?
Why
It could not at-
tract the C.B.S. affiliates at those hours, and its management must have deduced
that broadcasting at these hours could not be undertaken profitably with a small
roster of stations.
Until sufficient broadcasting outlets are available to a potential entrant
to allow it to reach a very large percentage of the nation's television homes,
market power will continue to reside in the hands of the existing three networks.
This market power will result in homogeneity of program fare as long as the
three networks are free to seek maximum profits, but attempts by the F.C.C. to
limit the profit potential of networks is not necessarily conducive to improved
program quality or to greater diversity.
The Commission's Proposed Rule must be
examined carefully in this light, for the mere fact that it may lead to lower
network profits is not a sufficient reason for supporting it.
pp.
34. See "United Network Forced to Quit," Broadcasting , June 5,
34-36, ff., for a description of the difficulties faced by United.
1967,
35. At the time United faced a minimum charge of approximately $100,000
per week for five eight-hour periods of microwave interconnections. Since this
time, A.T.ST, has reduced the minimum daily requirement for leasing trunk lines,
but it is now seeking a large increase in rental rates.
21
IV.
An Evaluation of the F.C.C. Proposed Rule
If it is the F.C.C.
's
failure to provide sufficient outlets to most of
the nation's television homes which explains network concentration, and if this
concentration explains the lack of program diversity, why has the Commission
chosen to attack the route by which programs reach the networks rather than
examining means to increase the number of channels available to most viewers?
Will the F.C.C. 's Proposed Rule have the effect of increasing program diversity
and making the syndication markets more competitive?
A careful examination of
these questions is attempted in this section.
Program diversity on the networks .
The Commission's argument that net-
works' direct licensing of programs reduces the variety and "quality" of enter-
tainment programming derives largely from an illusion about the willingness of
advertisers to commit resources to inefficient promotion and from a rather
romantic and nostalgic view of times past.
The Second Interim Report is laced
with allusions to the U.S. Steel Hour , Studio One
ALCOA Presents
.
,
Armstrong Circle Theater
and
,
These programs have disappeared in part because the advertisers
are no longer able to obtain network time for them, the Commission argues.
Net-
works, driven by their desire to maximize profits, are unwilling to schedule
such quality programs themselves.
The inescapable conclusion, according to the
Commission, is that diversity would be served by forcing networks to rely
heavily upon advertiser-supplied programs once again.
Unfortunately, the Commission's logic is faulty in a very elementary way.
If advertisers wish to subsidize "quality" programming through much higher costs
per viewer reached, why hav
they not made this demand felt from among the net-
36. Much of the Second Interim Report is devoted to summaries of the
testimony obtained from advertisers, producers, directors, and actors. Not
surprisingly, many of these individuals placed the responsibility for the de(Chapters X through XII.)
cline in "quality" programming upon the networks.
works' packager-licensed programs?
If these quality shows were remunerative
to the networks, would they, in their quest for profits, forego them for medi-
ocre westerns or situation comedies?
continue A. B.C. Stage '67
appealing?
,
For instance, why did A. B.C. fail to
a program which the Commission must have found most
The answer is to be found in advertisers' demand for the show
which averaged but 5.6 million homes, or 20 per cent of those watching network
television at the time, in the fall of 1966.
37
No large advertiser demon-
strated its desire to support such a program by defraying the networks
tunity cost, not U.S. Steel, Armstrong Cork Company, nor ALCOA.
•
oppor-
There are many
similar examples which could be cited, and they lead to the inescapable conclusion
that advertisers are unwilling to pay perhaps $6 or $7 per thousand homes per
commercial minute for "quality" entertainment when more common shows are available at $3.50 to $4 per thousand.
But if the Commission is correct in its assertion that the number of
truly fine dramatic shows has accompanied the decline in advertiser-supplied
programs, how can such a trend be explained?
in the opportunity costs of the medium.
The answer would appear to lie
It was not until 1953 that half of the
nation's homes were equipped with a receiver.
38
At present more than 95 per cent
have at least one set and another 30 per cent have at least two sets,
number of networks has remained at three.
39
but the
As a result, the value of network
time has increased enormously, and this fact in combination with increasing pro-
duction costs makes philanthropy a very costly policy for an advertiser on
today's network television.
In the early 1950s such firms as U.S. Steel, ALCOA,
37. American Research Bureau, Network Television Target Audience ,
1966
Interestingly, the F.C.C. has acknowledged that this program, though of
(P$F Radio
commendable quality, resulted in a net loss of $4 million to A. B.C.
Regulation Reports, 10 RR 2d 289 ff.)
issues.
38. Television Digest, Television Factbook , 1968.
39.
Ibid.
23
and Reynolds Metals were among the nation's largest television advertisers,
but they have reduced their network advertising in recent years,
40
perhaps
realizing that nationwide network advertisements, purchased at a cost of
$30,000 to $60,000 per minute, are not the most efficient means of promoting
the sale of metal producers' goods.
There may be a germ of validity in the Commission's charge, however.
While advertiser-supplied shows could not reach the networks without network
scrutiny and approval, networks did not oversee every production when advertisers supplied programs.
Producers, writers, and actors are understandably
uneasy about the current trend, for networks are likely to participate more in
creative details when purchasing programs directly.
Since there are but three
networks, this possibility of informal censorship is a serious matter.
But no one should be deluded into thinking that the networks would
agree to exhibit an advertiser-licensed program which promised to attract sig-
nificantly less than one-third of the total network audience during prime time.
The mere exhibition of such a series would reduce the value of all succeeding
shows on that network during the same evening because of the carryover effect
of audiences in television.
Once a viewer turns from a given network's pro-
gram, it is difficult to rouse him back to that network for the rest of the
As a result, advertisers would not only have to contribute more re-
evening.
sources per viewer in order to place a "quality" show on a network, but they
would have to promise the network greater net revenue than it could obtain from
a typical western or situation comedy in order to compensate it for later lost
revenues.
It is for this reason that networks usually offer their sacrifices
to the F.C.C.'s appetite for public service programming during the last hour
In 1968, these three firms spent only $8.3 million -- or less than the
cost of the sponsorship of a single hour series for a season -- on all television.
Network expenditures were $5.9 million. Only Reynolds, accounting for $6.4 million
of the above total, shows an increase in television advertising in recent years.
U.S. Steel has completely abandoned this medium, and ALCOA spent but $1.9 million
(Television Bureau of Advertising, 1968 Spot TV Expenditures and TVB News ,
in 1968.
40.
February 24, 1969.)
24
of prime time.
Competition in program supply
.
Even more puzzling is the Commission's
fear that program production is becoming concentrated as a result of networks'
control of program brokerage.
In its Notice , the Commission concludes:
"...it is not desirable for so few entities (the networks)
to have such a degree of power with respect to what the
American public may see and hear over so many television
stations.
...This intense concentration of power decreases
the competitive opportunity for independent program producers,
„41
At another point, the Commission cites the decline of new pilot films as a
sign of decreasing competition in program supply due to the difficulty in dealing with network monopoly power.
42
'
The Commission fears that a decline in the
number of producers will reduce syndication programming, the staple of independent
new UHF stations.
Given the networks' purchasing power and their insistence upon
syndication rights and/or profit shares, independent producers' viability is
threatened.
Without anticipating examination of the question of bargaining for
syndication interest in network programs, it might be useful to examine market
According to the Little
concentration in the program-supplying industry.
Report , the number of producers of television programs increased from 91 to 108
in the period 1958-67 but receded to 97 in 1968.
43
grams have declined from 75 to 57 in that period.
Broadcasting ,
I
'
Producers of network pro-
Utilizing data derived from
find that the number of suppliers of prime-time network enter-
tainment programs declined from 54 to 49 from 1958 to 1967, but
interpret this reduction as an omen of declining competition.
45
I
would hardly
Moreover,
rarely does one supplier have as much as 10 per cent of the market in a given
41. Notice of Proposed Rule-Making , p. 18. (Emphasis supplied.)
42. Ibid . , p. 13.
43. Little Report (1969), p. 94.
44. Ibid
45. It is impossible to detect joint ownership or control of program-supplying
companies; therefore, I cannot assert that these data reflect the number of independent
sources of prime-time programs.
It should be noted, however, that the most important
.
reason for the observed decline is the increasing average length of programs.
were fewer program series in 1967 than in 1958.
There
year and these market shares are extremely volatile over time.
Such a simple measurement of producer activity can be quite misleading
since the sources of programming for network and syndication markets are those
which supply films for any of a variety of entertainment media.
A production
organization is merely a collection of various creative agents, and this organization need not have any sizable investment in tangible production capacity as that
required for manufacturing firms.
Companies may be organized for producing a
single motion picture or television series in a few months and expire shortly
after the completion of this single task.
Production facilities can be leased
from a wide variety of sources in any of a number of countries for the particular
property to be filmed or taped.
Creative inputs exist and thrive regardless of
whether the network licenses its product directly or relies upon advertisersupplied programs.
The number of hours of programming produced for television
is not so much a function of network decisions as of F.C.C. decisions regarding
the number and distribution of licensed stations, CATV systems, and pay-TV
outlets.
But even if the degree of competition among producers were a function
of some parameter which the network can control, why should any network wish to
concentrate its purchases among a few producers or to act in such a way as to
drive others from the industry?
Does the Commission seriously believe that the
networks would create market power -- even if that were possible -- among the
suppliers of their most important input?
Network participation in syndication revenues
portance in the F.C.C.
's
.
Perhaps of greater im-
complaint is the assertion that networks use their
buying power to obtain rights and profit shares in program syndication.
Second Interim Report , the Commission's Office of Network Study argues:
In the
.
26
"Despite th(e) assertions by network managers, it seems
unlikely that their financial or proprietary interest in a
program series may not, in many cases, be a determining
factor in its choice for network exhibition. .But, assuming
normal commercial motives on the part of network managers,
this in and of itself might dictate an interested choice by
network managers between programs of apparently similar
worth as schedule components.
It would seem... that strong
economic motives exist to impel the choice by network managers of those programs for exhibition in which they have
acquired financial interests, the ultimate value of which may
be determined or enhanced merely by such a choice. "46
.
The Commission's Notice makes the same argument:
"While it has been contended that this (financial) interest is not a substantial factor in program choice, it
must be recognized that financial participation by network
corporations in any proposed program may well be the decisive factor in its selection for network exhibition. "47
The argument is a familiar one to students of industrial organization:
firms
with market power use their power, often through tying arrangements, to obtain
power in an adjacent market.
But the charge is even more extreme, for the net-
work company not only insists upon "financial participation" in programs which
it exhibits, but apparently succeeds in obtaining these interests at little or
no cost.
The Commission states that:
"Direct sale to sponsors had economic advantages for independent producers. Sponsors only occasionally acquired
or shared in syndication, foreign sales or other subsidiary
rights. These rights usually were retained by independent
producers and constituted valuable commercial assets which
contributed to their economic stability and viability. The
importance of the retention of these rights to the financial
stability of independent producers is supported by the testimony of producers that in many, if not most, instances
they do not recover their initial production costs from the
network run of a program series but must look to syndication
and foreign sales to 'make them whole' and to show a profit."
.„
Thus, the Commission is proposing to allow these producers to sell a
large share of their output to advertisers who generally forswear syndication
interests and rescue the suppliers from their impoverishment.
46. Op cit , p. 741.
47. Notice of Proposed Rule-Making , p. 18,
48. Ibid., p. 7.
.
)
11
The danger of network ownership of residual rights is thus seen to
derive from two quite different phenomena:
1.
Networks will gain control of an important source of supply of proSince the Commission is relying upon the
grams for struggling U.H.F. stations.
proliferation of these U.H.F. outlets to provide the basis for the entry of new
networks, it sees this control over the supply of syndicated programming as a
threat to future network competition.
2.
49
The networks* insistence upon syndication interests is leading to
the financial ruin of program suppliers -- the very backbone of program diversity.
If networks are not prohibited from exerting their monopsony power in dealing
with program suppliers, viewers will suffer a further reduction in quality pro-
gramming as suppliers are forced to exit from the industry.
1.
The control of syndication programs
.
This aspect of the Commission's
charge requires serious consideration, for if networks are gaining monopoly
power in syndication markets, it is possible that the number of U.H.F. stations
able to operate profitably will decline.
But once more the imprecision of the
Commission's analysis leads it to unsupported conclusions.
As mentioned above, syndication interests take two forms -- distribution
rights and profit shares.
Networks claim that they do not obtain distribution
rights in cases in which the program supplier has a distribution subsidiary,
but exceptions do exist.
Nevertheless, the three networks' degree of market
control is hardly staggering.
The three network companies have obtained distri-
bution rights in only 28 per cent of their prime-time entertainment -series hours
in the past ten years.
As a result, the networks have accounted for less than
one-fourth of all syndication sales in this period.
If the Commission believes
49. Perhaps the recent change in F.C.C. policy toward cable television
systems will change this emphasis.
(See fn. 18, supra
.
50. Little Report
(1969), p. 46.
28
that this degree of market concentration is likely to lead to a reduction in
syndicated-program supply, it might consider acting to curtail network distribution activity, but this does not imply that network ownership of profit shares
need also be reduced.
The ownership of profit shares entitles networks to share in the residual gain from program syndication.
There is no evidence that this investment
allows the networks or other participants to exert control over the distribution
of these programs.
Since the average network profit share is considerably less
than 50 per cent for those series in which it has interests, it seems unlikely
that such control could be exercised.
Nonnetwork distributors exist in large
numbers, and they are not likely to be dissuaded from competing simply because
the networks are entitled to a share of residual profits.
From all appearances, the syndication market appears to be quite competitive.
In addition to the old network series, there are "first-run" syndication
programs which are distributed by perhaps forty or fifty firms.
At least twelve
distributors -- including the network subsidiaries -- actively compete in the
sales of old network series.
As long as networks avoid accumulating the distri-
bution rights in a large share of their prime-time programs, they will not gain
market power in syndication.
It is clear that they realize this and that their
fear of antitrust action nas led them to be cautious in this respect.
2.
Program suppliers and network bargaining power
.
The Commission's
notion that the networks exert monopsony power to a degree which imperils the
financial stability of their suppliers is a curious one for several reasons.
First, it seems contradictory to observe that the number of independently-pro-
duced television series is increasing at a time when networks are setting prices
51. The Commission may well be acting against the interests of program
suppliers if it refuses to allow networks to include profit shares in their portfolios of income -earning assets.
This restriction might restrict the ability of
small program suppliers to defray some of the risk of producing programs by selling future, risky returns for guaranteed present payments.
too low for suppliers to cover their costs.
Second, it is unlikely that net-
works would wish to see the number of program suppliers decline, given the
importance of maintaining an uninterrupted flow of programs for exhibition.
Finally, it is not clear how monopsony power can be exerted without reducing
the quantity of programs demanded, but such a reduction has not occurred.
Elsewhere,
I
52
have argued that networks do not exert monopsony power in
the manner charged by the Commission.
53
They do not force suppliers to cede
them syndication interests; they simply buy them by paying higher costs per program.
An econometric examination of differences in program prices provides evi-
dence that the value of distribution rights plus a 100 per cent profit share is
between 13 and 17 per cent of the total contract price for first-run episodes
of a series on a network.
If networks pay suppliers for their syndication interests, what will be
the effect of prohibiting such investments?
Clearly, it will be to lower the
prices paid by networks to producers for their programs.
change in network program decisions?
Will it lead to a
The Commission charges that networks cur-
rently favor programs in which they (the networks) own syndication interests --
apparently on the mistaken notion that these interests are obtained at prices
below their present value.
Can such a "preference" be proven?
The Commission
only asserts the hypothesis; it does not attempt to substantiate it.
It is
possible to test this theory from evidence currently available to the Office of
Network Study.
52. The number of regularly-scheduled entertainment series hours in prime
time has been increasing since 1958 -- rising from 69 and 1/4 hours to 74 hours in
Some of these hours (approxi1965, but dropping slightly to 73 and 1/2 in 1968.
mately 20 per cent) are now given to exhibition of motion pictures, but these films
are bought from a subset of the entire program-supplying industry. Moreover, some
of these motion pictures are produced for initial exhibition on television, and
other prime-time series are similar to motion pictures in length and artistic nature,
53. Robert W. Crandall, "The Supply Price of Programs to Television Networks
and the Cost of Residual Rights," Bell Journal of Communications, forthcoming issue.
30
Each fall, the networks launch a new season of prime-time programs.
Some of these programs are retained from previous years while others are new
series.
With the data on each program series which the networks submit to the
F.C.C., it should be possible to discern if ownership of syndication interests
affects the networks' retention decisions.
54
These data include:
XI = average number of homes viewing the program during the
year, expressed as a percentage of total television homes.
X2 = a dummy variable indicating network ownership of distribution rights. X2 = 1 if the network owns the foreign
or domestic distribution right or both.
X2 =
if the
network owns neither distribution right.
X3 = the percentage of profits from syndication ("profit
shares") which accrue to the network company.
Observations are available for all regularly-scheduled prime-time series
scheduled for exhibition in the November Composite Week in 1900-65.
divided the programs for each year into two categories:
I
have
"retained" -- those
programs which were continued into the next fall season; and "dropped" -- those
programs dropped by the network before the next year's fall season.
The purpose
of the statistical analysis which follows is to detect the effect of X2 (distri-
bution rights) and X3 (profit shares) upon program-series survival.
If the
Commission's hypothesis is valid, a program is more likely to appear in the
"retained" category if X2
=
1
or if X3 is large than if X2 =
or X3 is small
or zero.
The first statistical test utilized is simple analysis of variance in
order to discover if an hypothesis of equal average values of X3 for retained
and dropped series may be rejected.
The test is performed upon each network's
data for each year (18 comparisons) and for the three networks' pooled data for
each year (6 comparisons).
Only the latter results are reproduced in Table
1;
the individual -network results appear in the Appendix.
54. These data were supplied to the author by the networks for the purpose of this study since information on program ownership is proprietary in nature
and could not be released by the F.C.C. to outside persons.
31
Table
1
Program Retention Decision
All Networks
Year
Mean XI
Retained
Series
Mean XI
Dropped
Series
Difference
F
Value
Mean X3
Retained
Series
Mean X3
Dropped
Series
Difference
Value
F
1960
21.45
15.07
6.4
40.6**
22.67
26.96
-4.3
0.7
1961
21.35
15.43
5.9
45.1**
17.58
27.40
-9.8
4.4*
1962
21.49
14.37
7.1
37.1**
21.01
24.93
-3.9
0.5
1963
22.36
15.10
7.3
40.8**
19.82
25.12
-5.3
1.9
1964
21.35
15.50
5.8
61.9**
20.95
23.01
-2.1
0.2
1965
21.29
15.29
6.0
64.6**
25.28
25.09
0.2
0.0
*
ciur
significant
at 5 per cent level.
** significant at
1
per cent level.
As might be expected, the average audience size for retained series is
significantly larger than that for series which were dropped the following season.
The results for profit shares are quite different, reflecting little significant
In only
difference in average network ownership in the two classes of programs.
one year (1961) is the network profit share difference significant at the five
per cent level, and in this instance the average network profit share in the
dropped series is larger.
The appropriate test for difference in distribution rights is the Chisquare test on the relative frequency of network ownership of domestic or foreign
distribution rights (or both) in the retained versus the dropped series
.
These
results appear in Table 2.
Table 2
Network Ownership of Distribution Rights
and the Program Retention Decision
Year
1960
1961
1962
1963
1964
1965
Proportion of Retained
Series in Which Networks
Owned Distribution Rights
AU
0.154
0.212
0.289
0.316
0.242
0.324
^significant at 5 per cent level
Networks
Proportion of Dropped Series
in Which Networks Owned
Distribution Rights
0.423
0.257
0.240
0.231
0.467
0.270
XT
4.52*
0.19
0.19
0.55
3.48
0.26
:
32
Once more, the results do not confirm the Commission's hypothesis.
In
only one of six comparisons is it possible to reject the hypothesis that the
relative frequency of network ownership of rights was equal for both categories,
and in that year (1960) the proportion of programs in which the networks owned
distribution rights was higher for dropped shows than for retained shows.
These results augur poorly for the hypothesis that network syndication
interests affect network retention decisions for a program of given audience
acceptance
In order to investigate this possibility,
.
it is necessary to intro-
duce all three variables into an analysis of covariance or to utilize discriminant analysis.
Our choice for this more sophisticated task is linear discriminant
analysis.
The object is to obtain from the data that function
Y = i^Xl + £ X2 + I X3
2
The X's have been
which discriminates best between retained and dropped series.
Vs
defined above and the
are coefficients to be estimated from the inverse of
The measure of the
the pooled dispersion matrix of the values of the X's.
discriminating power of each function is the Mahalanobis D
2
which is defined as
the square of the difference between the mean value of Y for retained shows, and
the mean value of Y for dropped shows
1)2
Tr
=
[
"
7d]
2
=
[V ri r
-
*V
V
=
l2
r
"
V V
r
+
l3
r
"
l3
*
d]
where the bars over the variable denote mean values and the subscripts r and d
denote retained and dropped series, respectively.
A test of the significance
of each discriminant requires the construction of the following F statistic:
F =
D
k,n
+n,-k-l
'
r
where
n
=
d
2
•
+n,-k-l)
n n, (n
r dv r d
k
(n +n
r
d)
(n +n -2)
r
d
number of observations in the sample of retained programs
n, = number of observations in the sample of dropped programs
k
=
number of variables in the discriminant.
33
In
order to measure the incremental contribution of X2 or X3, holding XI con-
stant, it is necessary first to estimate the discriminant which utilizes XI
only and then to add X2 and X3 separately and finally. together.
Table
3
contains
the results for the pooled three-network sample for each of the six years 1960-65.
Once more, it appears that the network retention decision is not influenced greatly by the ownership of distribution rights or profit shares.
In none
of the six years does X3 add significantly to the discriminating power of the
linear discriminant.
Distribution rights, X2, contribute significantly to this
discriminating power in three of six years -- two at the one per cent level and
one at the five per cent level -- but in every instance the sign of the coeffi-
cient of X2 is negative, implying that the program is more likely to be dropped,
ceteris paribus , if the network owns the distribution right.
Table
Linear Discriminant Analysis
-
3
Program Retention Decision
All Networks
Year
1960
Discriminant No.
1
2
3
•
0.4206
7.193
112.2**
XI
X2
0.4320
-1.77
10.446
160.4**
XI
0.4202
-0.0102
7.415
56.9**
10.552
53.1**
XI
X2
F
#
0.4316
-1.740
-0.006222
1
XI
0.553
10.730
167.4**
2
XI
X2
0.6065
-1.5227
13.402
112.1**
XI
0.5429
-0.01820
11.545
96.6**
X3
XI
0.5963
13.668
75.0**
3
4
X2
X3
1962
Value
XI
X3
1961
2
Coefficient
X3
4
Mahalanobis D
Variable
##
-1.398##
-0.01041
1
XI
0.3579
6.498
98.0**
2
XI
X2
0.3622
-0.2904
6.582
48.8**
6.560
48.6**
XI
X3
0.3588
0.001444
Table
Year
Discriminant No.
1962
1964
Coefficient
XI
0.3675
-0.4089
0.005573
2
XI
X2
0.3761
0.4610
7.678
58.3**
3
XI
0.3732
-0.006813
7.544
X3
4
XI
X2
X3
0.3710
0.8145
-0.01576
8.108
40.4**
1
XI
0.6915
16.371
257.3**
2
XI
0.7038
19.456
149.6**
X2
-1.3060##
XI
0.7036
-0.01826
17.259
133.4**
19.769
116.1**
X2
X3
0.7098
-1.211
-0.01124
1
XI
0.5986
12.888
238.4**
2
XI
X2
0.6102
0.7287
13.682
124.8**
XI
0.6048
0.0103
13.170
120.1**
0.6130
0.6668
0.006431
13.79
XI
X2
X3
##
-
33.3**
114.9**
XI
-
6.843
7.442
X5
8
Value
0.3756
4
-
F
XI
X3
**
Mahal anobis D
1
3
1905
(cont'd)
Variable
X2
X3
1965
3
significant at
1
82.7**
per cent level.
addition of new variable significantly increases discriminating power
at 5 per cent level
addition of new variable significantly increases discriminating power
at 1 per cent level
35
The individual network discriminant results, appearing in detail in Ap-
pendix Table 3, lead to a similar conclusion.
A summary of these results appears
as Table 4, below.
Table 4
Summary of Linear Discriminant Analysis
of the Retention Decision
Discriminant
Number
Variable Added
-
Individual Networks
Number of
Comparisons
Number of Significant Additions
to D
Positive
Coefficient
§5%
@1%
2
Negative
Coefficient
@5%
§1%
2
X2 to XI
18
2
2
3
X3 to XI
18
1
2
4
X2 to XI and X3
18
1
4
X3 to XI and X2
18
1
1
1
1
of 72 separate additions of either X2 or X3, only 6 provide a significant
increase in the explanatory power of the linear discriminant and possess the
positive sign required to support the Commission's hypothesis.
comparisons provide a significant increase in D
2
Six additional
but with a negative coefficient.
Sixty of the seventy-two additions provide no significant increase in ex-
planatory power, providing strong support for the theory that the value of
residual rights is capitalized into program purchase prices and that these
rights do not affect the retention decision.
In both the pooled and individual -network results, the coefficients of
X2 and X3 are more frequently negative than positive.
This slight inverse
relationship between syndication interests and the probability of retention, particularly for X2 in the early years of the sample, is
.
36
somewhat puzzling.
A possible explanation might lie in the magnitude of the
network's investment in these rights as reflected in the prices per network
episode paid to suppliers.
The value of the distribution right in a marginal
program may not be sufficiently large for the network to view continued investment in the program series as an attractive alternative.
In the
period, the profits from distribution fees were quite small,
*'
TJ60-63
and the present
value of distributing a mediocre series in syndication markets might have been
less than the cost of these rights which was built into the networks' program
payments
The discriminant analysis should be rather conclusive evidence that
network ownership of profit shares or distribution rights did not increase the
likelihood of annual renewal of a program series.
This is not surprising if
networks pay for tnese syndication interests through higher initial program prices.
A program of average success on the network and in which the network
lias
syndi-
cation interests is not likely to be more attractive than another program of equal
success in which the network has no interests because the former's cost per episode shown on the network is likely to be higher than that of the latter.
Indeed,
the above results show that in some instances ownership of the syndication right
may upon occasion increase the chance of a program's rejection.
The Commission's
theory of supplier exploitation simply cannot be supported from the evidence.
Summary
.
The analysis contained in this Section leads to the conclusion
that the F.C.C.'s Proposed Rule is based upon a number of misconceptions.
The
Commission's hypotheses which were constructed to explain changes in programming
practices in the past decade have led it to incorrect prescriptions for alleviating
poor performance in network broadcasting.
The next Section is devoted to providing
an alternative hypothesis to explain changes in programming practice.
55. See my "The Supply Price of Programs to Television Networks and the
Cost of Residual Rights," toe, cit., for discussion of this point.
37
V
An Alternative Hypothesis
Whether the F.C.C errs or not in its prescription for improving competition and variety in program supply, it cannot be doubted that the Proposed
Rule will have adverse effects upon the networks.
The networks express a concern that no one else could perform the
program-brokerage function, described in Section
this proposition is probably not correct.
I
In fact,
above.
Taken literally,
it is technically possible
for intermediaries to provide program brokerage, assembling programs for adver-
tisers and placing these programs on the networks, but such a development would
undoubtedly create a variety of problems.
Program brokerage could not be provided by a large number of firms.
Fifty per cent of the networks' prime-time programs amounts to only 35 hours
of entertainment series, and it is unlikely that the economies of scale are so
low as to permit a large number of firms to act as brokers for so few hours of
programming.
As a result, a bilateral oligopoly would develop in which networks
might find themselves paying considerably greater fees for brokerage than they
now encounter internally.
Moreover, the increased difficulty in arranging long-
term contracts with brokers, providing the networks with sufficient latitude to
proscribe program matter which is deemed not to be in the public interest, to
move programs from one time period to another, and to cancel programs which are
reducing audiences in subsequent programs, argues against such a prescription.
It seems quite clear that the development of packager-licensed program-
ming derived from the desire of networks to gain the additional flexibility in
the choice of programs and the sale of commercial time to advertisers provided by
their integration into program brokerage.
This would appear to be the key to
understanding a phenomenon which the Commission believes to be anticompetitive.
38
Program brokerage docs not provide an accretion to the networks
it
merely all ows them to benefit more handsomely from this power
To demonstrate the above hypothesis, we require
a
market power
'
;
.
model of advertiser
demand for commercial television time which embraces different means of purchasing advertising.
Recalling our discussion of Section III above, we may ask if network
profits are related to tne nature in which the network obtains its programming.
Will packager-licensing lead to larger profits than the reliance upon advertiser-
supplied programs?
'Die
answer would appear to be that the source of programs
is an important determinant of network profitability for a number of reasons.
I
P rogram
1.
In general, there are strong reasons for believing
choice.
that a network with monopoly power will not offer its advertisers the same menu
of programs as they would choose if they were supplying programs to the network.
A monopolistic network does not necessarily restrict
tiie
number of programs offered
to advertisers, but if the number of hours in prime time is a binding constraint,
it will reduce the
number of viewers offered during tnese hours through its choice
Monopolistic output restriction takes the form of reducing the
of programs.
number of viewer iiours
,
the network's principal output.
For packager-licensed programs, the network monopolist will choose pro-
grams in each program hour which satisfy the following condition:
(Wapp)
CD
where dA/dp
•
(mr)
=
i
represents the additional audience obtained from the last dollar
spent upon the hour of programming, MR is the incremental revenue obtained from
equals the program price.
the sale of another viewer hour, and p
This
P
result merely states
tiiat
the marginal value of the last dollar spent upon
assumed that for any particular type of programming, a network
faces an audience-supply schedule which is a function of its expenditure on program series.
It is also assumed that the advertisers demand, and networks supply,
a flow of homogeneous viewers -- an assumption which is examined below.
56.
It is
39
programming must be equal to one dollar.
As long as
2
9
A/9p
2
is less than
zero -- indicating diminishing marginal productivity of program expense -this condition will assure maximum profits for the network.
MR,
is obtained from the demand curve facing the network
Marginal revenue,
monopolist -- a
demand curve which is simply the horizontal summation of advertisers' marginalvalue -product curves for advertising, measured in viewer-hours per week.
The advertiser will always equate his marginal value product of adver-
tising to the price of viewer hours.
choose p
such that (p
If he supplies his own program, he will
minimum, where p
+ p )/A is a
t
is the hourly rate
charged by the network monopolist for access to its affiliated stations.
This
simply argues that an advertiser chooses his program in order to minimize his
cost per viewer, which requires that:
(2)
where 9A/9p
MVP
A
=
P t )/(3A/9p ).(p + p t ) = l/(9A/8p )
p
p
p
(p
p
is once again the incremental productivity of another dollar spent
on the program in terms of viewer hours,
57
'
and MVP. is the advertiser's
Under packager licensing, the advertiser
marginal value product of advertising.
will choose to buy commercial minutes up to the point where:
MVP
(3)
A
= p
A
where p. is the price per viewer hour connoted by the network's charge for com-
mercial minutes.
From
(1)
l/(9A/9p
(4)
and (3) above, we see that for packager licensing:
=
)
MR
<
MVP
A
Therefore, a network monopolist will choose
since the network has market power.
less costly programs for sale to advertisers than would the advertisers if they
were supplying their own series.
Regardless of the time rate chosen by the network,
57. This result is obtained by differentiating (p
with respect to p
3[(P + P t )/A]/9p
=
[A-(p +p
p
choice; therefore, A-(p +p
t
)
)
(9A/9p )]/A
(3A/9p
)
=
= 0.
But A
2
t
for any program
and A = (p +p ) (9A/9p
t
substituting this value for A in (p +p )/A yields (2).
partially
This results in:
and setting the result equal to zero.
2
+ p )/A
t
)
for an optimum,
40
advertisers will choose more costly programs, reacting to changes in the
network's time rate by increasing or decreasing the number of programs they
choose to bring to the network for exhibition.
It is possible to represent this result graphically by drawing those
schedules which affect the programming decision in the incremental hour and
the effect of this decision upon intramarginal hours of network broadcasting.
In Figure 2 the relevant cost and demand schedules for the incremental hour of
network broadcasting are drawn.
The ZMVP. curve is the aggregate demand schedule
for viewer hours in the incremental hour ; therefore, the network's marginal reve-
nue function is not the curve marginal to this ZMVP. curve, but rather it is the
rate of change of total advertising expenditures for all hours during the week.
The network licensing its own programs will set l/(9A/3p
)
-- the marginal
cost of audience -- equal to MR, choosing a program of audience size A* whose
cost per viewer is (p /A)*.
If advertisers license programs they will tend to
choose a program for which MVP. is equal to l/(3A/3p ), and the network's optimal
a
p
time charge, p
,
is found by setting p /A** equal to p A ** minus
Figure
(p /A)**.
If it
2
The Program Choice for a Given Hour of Network Broadcasting
1
*AA>
i*
p
~
Pp + Pt
A
MVP A
A
**
Pp/ A
(P„/A)*
(Pp/A)*
A= Audience
41
chooses a higher time charge, the advertisers will find that the total cost of
the optimal program of audience A**, read from the curve
(p
+p.)/A, will be
greater than its value, and they will choose not to supply the program for this
incremental hour.
Instead, they will optimize over the first n-1 hours of
prime time, increasing their expenditures on programs, p
,
in these hours. They
will do so simply because the incremental cost of additional viewer hours will
be less in these n-1 hours than in the nth hour, given the network's time charge,
Advertisers supplying their own programs therefore choose different
program "quality" than the network would if it were free to license programs
directly.
The cost per week to the network of this freedom of advertiser pro-
gram choice can be measured from Figure 2.
incremental cost schedule, l/(3p /3A)
,
It is the shaded area between the
and the network incremental revenue
schedule, MR, multiplied by the number of hours of programming per week.
Note
that this cost would be zero if the network possessed no market power.
There is an additional factor which will lead a network licensing its
own programs to restrict audience size during the incremental hour even further
than suggested by the above analysis.
The addition to the network's total audi-
ence occasioned by programming during this incremental hour is not simply the
number of viewers during that hour, but some smaller magnitude depending upon
the degree to which viewers restrict their viewing in the n-1 hours of the week
due to the nth program hour.
This difference depends upon the "shiftability"
of viewer preferences, i.e. ..the degree to which viewers are able or willing to
shift the time periods during which they choose to watch particular types of
programs.
58
In terms of Figure 2, this simply means that for any program
broadcast in the incremental hour, 9A/3p
is likely to be lower for the network
58. See Peter Steiner, "Program Patterns and Preferences and the Workability of Competition in Radio Broadcasting," Quarterly Journal of Economics ,
May 1952, pp. 194-223, for a thorough discussion of the importance of shiftability
in program decisions.
42
than for the advertiser who licenses but one program directly.
l/(3A/3p
pJ
As a result,
for the network will be above and to the left of the corresponding
curve for the advertiser supplying his own program.
These results may appear surprising at first, but this is perhaps due
to a tendency to view the network's output as program hours when it is in fact
vi ewer hours.
Under a system of advertiser-supplied programs, the network
cannot affect the incremental cost of viewers to the advertisers; it can only
increase the average cost of these audiences.
Advertisers will always be
free to choose programs efficiently, and this will be against the network's
best interests.
Extension of this model to a three-firm oligopoly requires, as always,
some assumptions about the reaction function of each network.
The nature of
the final program-price equilibrium depends crucially upon such assumptions.
If
networks behave as Cournot oligopolists -- choosing program prices as if their
rivals' choices were fixed -- they will undoubtedly reach an equilibrium at
which audiences are larger and program prices greater than under simple monopoly, but the equilibrium will find smaller outlays for programs than those
cnosen by advertisers supplying their own programs.
As the number of networks
increases, the equilibrium will resemble a competitive one and a similar pro-
gram structure will emerge under packager licensing or advertiser supplying of
programs.
This model leads to a quite different conclusion from that obtained by
the Commission in its analysis of programming practices.
Networks offer programs
which are less expensive and appeal to somewhat smaller audiences than would
advertisers supplying their own program series.
Of course, this difference may
be based upon different implicit models of advertiser behavior.
I
see advertisers
as seeking the greatest exposure per dollar spent; the Commission argues that some
A full discussion of this problem is not possible in this paper.
'Television Network Competition and Program Diversity" (unpublisned
manuscript) for such a discussion.
5 J.
l
Sec my
43
advertisers may be seeking to promote an "image" by supplying programs which appeal to a narrow audience at a very high cost per viewer.
But the two models
may not be exclusive -- advertisers seeking "image" will still seek to mini-
mize their cost per viewer of a given type.
A network monopolist, licensing
its own programs, can just as easily contract the supply of a select audience to such
advertisers as the supply of Joneses to the rest.
2.
argument.
Augmentation of demand
.
This is a very simple, but most important
Networks have moved to packager licensing and the dispensing of com-
mercial minutes in order to increase the market
ing.
demand for network advertis-
Under a system of advertiser supply, advertisers usually were required to
subscribe at least half of the cost of a half-hour series for one season.
Al-
ternate sponsorship of a new half -hour series would cost at least $2 million
current prices, a total far greater than the advertising budget of many current
purchasers of network advertisers.
The old system was simply rationing out too
many small buyers of network time.
Of 439 network advertisers in 1968, 247
spent less than $1 million in the medium during the year.
6U
It is the
reduction
of this minimum investment from, say, $2 million to as little as $30,000 per
year which is essential to maximum network profits; therefore, someone must serve
as a
program broker.
It was argued above that it seems unlikely that an independ-
ent brokerage market could develop downstream from a concentrated network market
or that an independent brokerage market would supply its services as efficiently
for networks as a vertically-integrated firm.
In addition, the
opportunity afforded advertisers to disperse their com-
mercial announcements among many different prime-time series
the market demand for viewer hours in another manner.
lias
probably increased
The marginal value product
of advertising is likely to decline more rapidly the more an advertiser is con-
strained to address the same audience.
Since television programs are known to
60. Television Bureau of Advertising, TVB News, February 24, 1969.
44
have devoted viewers who watcii each episode without fail, requiring advertisers
to concentrate their purchases in a single, or at most a few programs will
probably lower the MVP. schedule in Figure
2.
Through program brokerage, net-
works are able to offer more dispersed patterns of commercial announcements
to their customers and to profit accordingly.
Finally, the ability to dispense commercial minutes one at a time or in
large packages provides greater flexibility in pricing if advertisers possess
different price elasticities of demand.
Those whose demand is relatively price-
inelastic may be separated from those with more price-elastic demand.
sucli
While
discrimination has not been proved by critics of the industry, this oppor-
tunity is certainly eniianced by the networks' own performance of program brokerage.
These considerations would seem to be the most important explanations
for network opposition to the Commission's Proposed Rule.
Requiring that networks
foreswear syndication revenues is not likely to reduce network profits appreciably, but requiring that they depend upon individual entities to act as dealers
and brokers of programs for network exhibition may be quite costly to the networks without a corresponding gain in economic welfare.
3.
Risk bearing
.
The development of network program brokerage may well
be a reflection that networks are better able to spread the risks of investing
in costly entertainment series.
The discussion in this paper has tended to treat
program purchases as if they were effected in a spot market, one episode at a
time.
In fact program purchasers, whether they are networks or advertisers, must
invest in a number of episodes of a program series before the series appears on
61. Recent evidence presented by David Blank ("Television Advertising:
The Great Discount Illusion, or Tonypandy Revisited," Journal of Business January
1968, pp. 10-38) and John L. Peterman ("The Clorox Case and Television Rate Structures," Journal of Law and Economics , October 1968, pp. 321-422) lends support to
the notion that networks do not discriminate in favor of large advertisers, but this
It may be that
is not conclusive proof of the absence of price discrimination.
price-elasticity of advertiser demand is not related to advertiser size.
,
45
network television.
Commitments of 13, 26, or 59 weeks are not uncommon, and
buyers must pay penalty fees for cancellations short of such commitments.
Since the success of a program in its initial network exhibition and
in
future syndication is a function of its position in the network schedule, an
advertiser supplying but one program may face a greater variance in
to his investment than a network which may pool
series.
tiic
returns
this risk over a number of
Moreover, the network has greater latitude in shifting programs among
time periods in order to achieve the best results for its array of programs
when it licenses these programs directly.
Whether these considerations lead
to a significant reduction in the risk of supplying programs cannot be determined
in this paper, but this possibility does exist.
Any reduction in the risk as-
sociated with program procurement obviously lowers to supply price of programs
and increases network net revenues for a given array of program choices.
In summary,
it is suggested that the networks have turned to direct
licensing of their own programs in order to increase the net demand for prime
time.
While there are reasons to believe that this backward integration has
led to a change in program patterns, this change has not nad the adverse effects
upon program diversity or quality which the F.C.C. believes.
VI.
Conclusions
This analysis of network program procurement practices has been rather
general in nature, requiring the separate presentation of a number of issues
raised by recent changes in network practices.
In part, this generality is oc-
casioned by the necessity of describing a market structure which is little understood by economists, but much of the discussion has been required in order to
address the far-ranging criticisms raised by the F.C.C.
Unfortunately, the
Proposed Rule -- outlinedin Section II -- is thought by some to be a palliative
46
for countless evils of network broadcasting.
The conclusions reached in this
paper are that such a rule -- requiring networks to limit direct licensing of
program series and to foreswear domestic syndication interests in programs
--
will have little beneficial effect upon the parameters which the Commission
deems important:
network program diversity, competition among program suppliers,
and competition in syndication.
The most likely effect of the Rule will be to lower network profits by
allowing advertisers to choose more expensive programs and by making more cumbersome the performance of program brokerage.
The effect of such changes upon
economic welfare would be difficult to assess, but it is a safe prediction that
network program schedules will continue to exhibit little diversity until networks are more numerous.
In this respect,
the Commission's recent decision to
encourage cable antenna television companies to originate their own programming
may be the single most important step along the road to greater competition in
network broadcasting and increasing diversity in programming.
#
#
#
#
#
•
1
<
1
<
1
<
(
(
i
1
<i
<
*
1
i
1
1
47
o
3
—
OQNOOH
O O
:3
to
>
IO ro
o3
~3
"5
<n
r-»
HI1HO
-t O CN —
3
Cr>
10
^H
O
H
T3
O —
1
i
CD
'J
»—
00
—
—
ai
<
1
4-1
i-t
1
1
IN
—<
~H
O
O
CN .0
o
rr>
rH
1
11 T3
C
o
— H
U
1
CU-H
d O
M CN
Q
ohm
a
(h
7.
to
X
i—
o
cO
8)
T
f\|
10
~J
O
r-|
1^
<N
(N
10 51
i—4
10 <N
1
1
O
tO
r~~
"1
r-~
o o o
i
1
-3 r^
ro -d
HO
tO
"*
*d
o
oO
h—11—4
oh
I—
-± co to rCO
\3 10
CO
n
OO
1
eo up
1
1
r-l
i^
<N cn
CT>
—
t-^
tO CN CO IO
.
cm co -r
CO
J>
O
to -t •* r^
r
o
— —
to
4
i
o
ri
o
i
O
1.0
—
io -t
cm r-j rj 10
to
—
3
o n
i
r-4
3
o
o
t^~
1
OH
O
tO —l ~J ~> f 1 :o
(0 CO
en
to
&4
o
o
o o o
o o — -h
rs to
~3 to
r-4
Cm
x
*
*
1
•cf
—
o
CO
o
<"
1
l.O
IO
tfl
O tO
c a
-H
w
o
o
co
^y
CO
O
— *
r-l
CO
4->
to
o "J
O
oo
to
t*-
Tf
CO .0
CN IN to
r\i
r--
'O
r—
to
to 00 ro <"•! r>
CO 31 f-- rsi -H CO
CO
o
—
f-^
O
cn to 10
r-H
rt
T—
rj
IN
— —
i
O O
CO
t)
'0 CO r-t
'/> "3 CO
3 t
o o —7i rj
—
4
--1
r-l
<
CJ
ct
o
H
r"«
co
-4
•H
ccj
O
e>
(-1
—1
r-i
to
rt
*=
tt
*t=
->
CD.
CM
5-<
^ o
o
•rH
XJ
™
ai
10
H
X
i-H
-j
co
M
a3
C^c
*2
—<
to
j->
O
o
3
0>
3»
*
*
*
*
*
O J OD
•cr
r^
o
i
*
0"l
LL,
*
*
*
*
*
o
LO CO
t3 —i CO
a
O
10
r^
—4 !N —i r-t
o
3
u
*
*
c^
—
o
*
*
10 CO
cm
*
*
O CO
m
T
*
*
o
*
o
^»
*
+J
O
00 10
CM r> <N rj T> —f
to
t~^ *—
—
—
o
cci
C"*
D
r
l
—i -r 'o
ceo t^ IO
HN^
l
t^-
c7i
O
to
CO
O
IO
—
(
> >
o
o
CD
u
o
O O :0
O
r-j
*& -t :Q
O
-rr
<+4
O
-o
<~i
CTi
C71
ri
"t tO
O
CO to
L0
*IHH
MN
O T cj
O
O M fO N O
tO Tt •— rj IO CO
io to io i-H -r io
l
CO *3 -t
o
IO t^
CD
u
U-i
O
r-C
O
cn,
IO
rJ
(0
—
X
ccj
rcj
O
~
T3
O
tO
a.
<i>
-r r» ai r-~ r^
r- r-^ TT> to r j
PU-H
o ^
-i-
rsj
to
O
LO
•*
f^
o
r~c~.
:o
r^
o
r-~
o
oo
-j-
r~^
io
r--
to
o
CN
i
—
t~- •o -H r>
C0 <N t^ ri r-l
l~-
r-«
cc;
'J
'J
U-4
.-I
M-<
-o
-r lo 'O -t io io
0)
^i
a
••o
o
r^-
J-l
co
-H
CO CO
-H
H
V)
-o
r4
•—I
X
1)
.
*
(
JD
CO
t_
to
0)
<u
S
•H -H
cc!
CO
J-
cd
J>
+J
o
" s
CO
CD
-r to T\ Ol LO
-T -t- —l
n
r-<
o
rj
7)
r^
O o en
— ^)
—
CD
—
"J to rr LO
T>
1
o
-o-
1>1
o o
Lo ^j
c»
to -t ^h -r •£>
~I CM IO IO CM
C^J
3
CO
tN
'-O
i
C^J
en C3
t
'1
c-j
IO
CI r4 'J IN CN CM
r
r
-r
o —
o o o o o
o
j ci ci o
o
3 a ^
CO
CO
o
'1
HX
-3
r—
"J
Ci.
r^
(
<;
u
CO
<D
>-
O
n ma
O
C7>
o
m m
-O
o
en
i
"-i
to
io
D
ri
r-t
i~j
io -r to
II
3
J.
10
*
*
48
Appendix Tabic
2
Retention Decision
Syndication Rights--Chi Square Test
Year
Proportion of
Retained Series
in which
Network Owned
Syndication
Rights
Proportion of
Dropped Series
in which
Network Owned
Syndication
Rights
Network
i%o
I9bl
1962
1963
1964
19u5
0.00
0.835
0.500
0.250
0.250
0.222
Difference
#1
-0.333*
O.lOO
0.146
0.107
-0.050
0.055
0.333
0.067
0.154
0.143
0.300
0.167
Network #2
11)00
E961
1962
1963
1964
1965
0.250
0.167
0.263
0.375
0.508
0.407
-0.350
-0.333
-0.070
0.250
-0.328
0.107
o.ooo
0.500
0.353
0.125
0.636
0.300
Ne twork #3
1960
1961
1962
1%3
1964
1965
0.300
0.444
0.533
0. 333
0.125
0.251
*
-0.117
0.144
-0.000
-0.031
-0.319
-0.213
0.417
300
0. 333
0.304
0.444
0.444
.
difference significant at
5
per cent level
«
Appendix Table 3
49
Retention Decision
Discriminant Analysis
Network ttl
1900
Varial >le
XI
"P*
0.2827
1.545**
XI
0.4701
X2
-6.910M
1901
0.4668
4.182**
0.0719
7.185**
0.315S
-0.ool4
2.134**
0.5340
0.2747
12.309**
0.795O
1.122
51.300**
0.5189
-1.405
5.489**
0.6909
0.9279
7.881**
0*
0.5098
0.004310
2.066**
T>*
0.4705
-b.909
-0.002358
19.125**
0.3209
-0.7544
O.OO0668
2.171*
X2
X5
190 j
2S. 998**
0.2830
-0.002957
1.557*
XI
1904
0.5335
12.245**
19. 097**
X3
1903
0.5075
2.051**
D*
XI
1902
0.5470
0.7900
0.4o77
U. 02432
4.4 09**
0.005O
0.009755
16.131**
0.7905
0.003040
29.041**
0.3493
-0.1374
0.04 044
10.154**
0.8030
1.239
-0.000517
31.460**
0.5487
-1.583
-0.01726
5.984**
0.6846
0.9357
0.0100
8.050**
0.03975ff
Network
7.34 0**
112
P1
0.0165
15.173**
0.9168
36.S53**
0.2152
1.553
0.3187
6.521**
0.9483
49.252**
0.4217
6.946**
D*
0.5962
-0.4985
15.535**
1.161
-4.287##
83.171
0.2342
-1.163
2.226
0.3084
0.7031
7.004**
0.9SO2
-1.718
58.383**
0.4476
1.2S6*
9.071**
O*
0.6084
-0.006127
15.261**
1.137
-0.1236##
97.984
0.2140
-0.0150
1.749
0.3244
-0.08012
7.701**
D*
0.5960
-0.4918
-0.000357
15.535**
XI
XI
u
XI
X3
XI
X2
X3
1.2371
-2.700
-0.09315
118.294**
0.2250
-1.584
0.01286
2.300
Network
75.702**
0.4239
0.02242
8.020**
-0.077S8#
11.340**
1.082
-0.3428
-0.06543
73.996**
0.4447
1.104
0.003218
9.135**
0.2988
3.217/.'
1
.
090
-0.07036#tf
//5
x
0.5475
20.310**
1 543
97.078**
0.4480
3.730**
0.O0O3
13.485**
0.8366
23.146**
1.544
136.984**
1.576
-0.724
99.621**
0.4496
0.1723
3.756**
0.7358
2.040tf
"D*
0.5406
-0.07231
20.825**
19.074**
0.3719
-0.2943
23.284**
1.559
-0.2545
137.52**
x
0.5521
-0.01370
21.655**
1.541
-0.01304
99.718**
0.4472
-0.01443
4.055**
0.6202
-0.03518
17.650**
0.8897
0.006688
23.272**
-0.2674
141 .9o2**
D*
0.5515
-0.04378
-0.01372
21.655**
1.566
-0.5402
-0.01473
100.816**
0.4517
0.5459
-0.02012
4.273**
3949
3.320##
-0.7459/^
38.713**
0.8603
-0.5197
0.009949
23.741**
0.03387
-0.02698
142.053**
XI
T>
XI
X2
XI
X3
P
XI
X2
X5
.
.
1
1
.553
.554
* discriminant significant at 5 per cent level
** discriminant significant at 1 per cent level
tt
additional variable increases discriminating power at 5 per cent level
## additional variable increases discriminating power at 1 per cent level
1
i
-
1
DitaJUo
HN 04-82
APR 1 7 '82
8gp24'fl£>
(jw-mjjb
DEC
5
19! 11
JSAR31 -Ijl
/VfN
2-tf
-jctn
•?
?
?r
i
1
SEC 2 7
7
MAT
^«
TO El
!
Mia
BY
2 9
'
....,
e
79
1992
_.
.
I
!
::
7~
Lib-26- -67
3
TOflO
DD3 1ST SE2
MIT LIBRARIES
TDflD
3
DD3
^5=)
34Q
MIT LIBRARIES
r
t. ".
3
TDflO
3 IQflD
DD3 TST 530
DQ3
T2fi
511
MlT LIBRARIES
3
TOAD DD3 TST SDh
MIT LIBRARIES
3
3
TDflD
DD3 TST
TOAD D03 TST
3t>5
4
MIT LIBRARIES
TOflQ
DD3 7Tb 072
53056
>otiomics